CN102202484B - Diagnosis method for heat radiation system - Google Patents

Abstract

The invention discloses a failure diagnosis method for the heat radiation system of electrical equipment. The method comprises the following steps: introducing a step signal in a stable work state of the electrical equipment; measuring the temperature of the heating element of the electrical equipment; carrying out signal processing on the obtained temperature value; comparing the result of signal processing with a predetermined threshold, wherein when the result is greater than the threshold, the heat radiation system is judged as abnormality which means failure. Therefore, according to the diagnosis method of the present invention, the work status of the heat radiation system is determined in advance, which facilitates the arrangement of maintenance and the repair of the heat radiation system in dependence on work status.

Description

Diagnosis method for heat radiation system

Technical field

The present invention relates to a kind ofly to the method for diagnosing for the cooling system of electric equipment, and relate in particular to and use straightforward procedure to detect cooling system fault, to prepare for maintenance and the renewal of cooling system.

Background technology

Along with the heat radiation increase of power electric equipment and reducing of global shape factor, it is more and more important that heat management becomes.

The temperature of the performance reliability of power electric equipment and life-span and the each device of equipment is inversely proportional to.Typically, the relation between reliability and the working temperature of silicon semiconductor device shows that temperature reduces reliability and the life-span increase corresponding to semiconductor device.ILS and the reliability that therefore, by the working temperature of control device effectively, can realize each device increase.Each device that thus, conventionally will guarantee this power electric equipment is worked under the threshold temperature of design.

Radiator by through be commonly used to hot surface from heater members by heat dissipation to surrounding environment, for example, air.In the present embodiment, take air as example, but can be other media, as oil.In order to improve the radiating efficiency of radiator, usual way is to increase radiator and cooling agent, the surface area that air contacts, or improve speed air flow with fan.

But along with increase service time, radiator can occur stopping up due to the deposition of dust etc., and fan also there will be aging, thereby the overall efficiency of cooling system is declined, can not guarantee that main electric part obtains fully cooling.And well-known, excess Temperature can cause electric part too fast aging, reduces the life-span of electric part, and likely causes electric part to break down.

For fear of electric part, because high temperature is damaged, in common electric equipment, for example, in frequency converter, at power model, place is provided with temperature sensor, when the measured temperature of temperature sensor exceedes a predetermined value, stops the work of frequency converter.

But owing to not having at present a kind of method to diagnose cooling system, thereby unpredictable when cooling system goes wrong, therefore, because overheated meeting causes extemporal shutdown, thereby cause irretrievable and unnecessary loss.

If for example, in textile industry application, may, due to the fiber in applied environment or filiform obstruction radiator or fan, radiating effect be declined, frequency converter be shut down.If can be detected and be known that in advance fan or cooling system go wrong by system, prompting client clears up or changes fan in advance, can avoid uncertain shutdown.Or after shutting down, also can find very soon the reason of shutdown, take very soon relevant treatment measure.(renewing fan or cleaning radiator)

Summary of the invention

In view of the above problems, the present invention has been proposed.Therefore, the object of the present invention is to provide a kind of diagnostic method of the cooling system for electric equipment, utilize the method, can to cooling system, carry out failure diagnosis at any time, and can preset according to deagnostic structure preventive maintenance time and the replacing time of cooling system.

In order to realize object of the present invention, according to the present invention, provide a kind of diagnostic method of inefficacy of the cooling system for electric equipment, comprise the following steps:

At described electric equipment, under stable state, to it, introduce a step signal;

Introducing after this step signal, measuring the temperature of the heater element of described electric equipment;

Obtained temperature value is carried out to signal processing; And

Result after treatment signal is compared with predetermined threshold, and when described result is greater than described threshold value, judge that described cooling system work is undesired.

Described stable state refers to that described electric equipment is operated under stable load and switching frequency, maintains stable working temperature.Described cooling system work is undesired comprises that the fan of this cooling system is aging, and fan failure or radiator are blocked.

Thus, by diagnostic method of the present invention, can determine in advance the working condition of cooling system, to realize the maintenance and repair that arrange cooling system according to working condition.

Preferably, described step signal is power loss step signal.Described power loss step signal is that switching frequency or the change load by changing described electric equipment realizes.

The step of obtained temperature value being carried out to signal processing comprises:

To obtained temperature value seeking time derivative; And

By obtained time-derivative by a coefficient k ₀remove, the derivative value of above-mentioned measured temperature when this coefficient is time t=0, that is: ${\mathrm{<figure-callout\; id="0"\; label="k"\; filenames="HSA00000040449900081.png"\; state="\{\{state\}\}">k</figure-callout>}}_0=\frac{d\&CenterDot;T_c\left(0\right)}{\mathrm{dt}}=\frac{P_{\mathrm{AV}}}{\mathrm{Ct}{h}_{\mathrm{ca}}}.$

Described threshold value is:

Wherein: Rth _cait is the thermal resistance value of cooling system under normal operation;

Cth is the thermal capacitance of cooling system; And

T is the testing time.

And testing time t is $t=\frac{{\mathrm{\&tau;}}_{\mathrm{<figure-callout\; id="4"\; label="ca"\; filenames="HSA00000040449900081.png"\; state="\{\{state\}\}">ca</figure-callout>}}}{4}$ Or larger, the thermal time constant that wherein τ is system.

This shows, in said method, removed the impact of power loss, improved the robustness of diagnostic method.

Accompanying drawing explanation

Below, with reference to accompanying drawing, describe in detail according to a preferred embodiment of the invention, still, it is to be noted that description is below only exemplary, and be not construed as limiting the invention, in figure:

Fig. 1 is the curve that represents the relation between the parameters such as thermal resistance, temperature rise;

Fig. 2 is illustrated in place, sea level, in the cooling system with fan, and thermal resistance and the relation curve of air velocity that flows through radiator;

Fig. 3 is illustrated in 2500 meters of height above sea level places, in the cooling system with fan, and the curve of the relation of thermal resistance and air velocity;

Fig. 4 is illustrated in place, sea level, the relation curve of thermal resistance and power loss in the cooling system that there is no fan;

Fig. 5 A is the view that the model of cooling system is shown to 5C;

Fig. 6 is the control chart of a kind of diagnostic method of illustrating that the applicant conceives;

Fig. 7 illustrates the improved control chart of above-mentioned diagnostic method;

Fig. 8 is the control chart illustrating according to the execution mode of diagnostic method of the present invention;

Fig. 9 and 10 is that explanation is according to the view of the robustness of the diagnostic method of embodiment of the present invention.

Embodiment

With reference to accompanying drawing, the preferred embodiment of the present invention is described below.At this, it is to be noted, although the present invention is described as an example of the power model of frequency converter example, the present invention is in no way limited to frequency converter, but can be applied to any use cooling system, carries out cooling occasion.

Before describing diagnosis method for heat radiation system of the present invention, first provide the definition of some parameters:

P: the gross power of heat radiation or rate of heat dissipation, or the power loss of electric equipment, unit is W, represents the heat speed that electric component dissipates in the course of the work.May also be referred to as the power loss of electric equipment, because the power of electric equipment institute loss will become heat, and this part heat needs the gross power of heat radiation just.In order to select to design radiator, should use maximum service rating to dissipate;

Tj: the maximum junction temperature of electric component, unit is degree Celsius, in typical electric equipment, admissible Tj from 120 degree in higher scope, for maximum working temperature, likely to 170 degree or higher;

Tc: the case temperature of electric component, unit is degree Celsius, because the case temperature of electric equipment depends on measuring position, it represents the maximum local temperature of housing conventionally;

Th: radiator temperature, unit is degree Celsius that this Parametric Representation is at the radiator maximum temperature of the position radiator of close electric component;

Ta: ambient temperature, unit is degree Celsius.

Utilize above-mentioned parameter, between two diverse locations, heat transference efficiency can be expressed as:

ΔT _ca＝Rth·P

ΔT _ca＝T _c-T _a

Wherein, Rth is thermal resistance.

Take the power model of frequency converter as example, Tj refers to the temperature at the chip place of power model, and Tc refers to the temperature at power model housing place, and Ta refers to the ambient temperature of the environment that frequency converter works.

τ: thermal time constant, adopts this parameter by homogeneous heating or when cooling at object.In this case, within preset time, from the object heat transmission of environment towards periphery, be proportional to the temperature difference of object in the middle of surrounding environment

τ＝Rth _ca·Cth _ca

Wherein Cth _cathermal mass, also referred to as thermal capacity, for radiator, this thermal capacity Cth _cait is the function of radiator material and radiator volume.

Fig. 1 is the curve that represents the relation between the parameters such as thermal resistance, temperature rise.The curve of Fig. 1 is the combination of two independent curves, and they are combined in Fig. 1.Suppose that the device that will be cooled is correctly installed, and with respect to airflow direction (Down-Up), radiator is orientated in normally used installation.Curve from lower-left to upper right represents the free convection curve of radiator temperature rise.Free convection phenomenon is nonlinear.

From curve left to bottom right, be the forced convertion curve of thermal resistance with respect to air velocity.In forced convertion, the linear ratio of Δ Tca and P, Rth thus _cairrelevant with P, and be only the function of flow velocity.

Another parameter that need to consider is height above sea level effect.Although the air themperature of indoor environment can be controlled and not be subject to the impact of height change conventionally, room air pressure can change along with height.Because a lot of frequency converters are arranged on different height above sea level places, need to consider: at higher height above sea level place, air pressure reduces and atmospheric density reduces, and the radiating efficiency of radiator can be affected thus, and following table 1 illustrates that height above sea level reduces parameter to sink-efficiency.Shown in table 1 at different altitude height place the reduction parameter to different radiators.For example, for the actual hot property of the height above sea level place radiator beyond determining on sea level, the thermal resistance value of reading from performance curve will be removed by these reduction factors before comparing with required thermal resistance.

Table 1: height above sea level reduction factor

Height above sea level m/ft	Reduction factor
			0, sea level	1.00
1000/3000	0.95
		1500/5000	0.90
2000/7000	0.86
		3000/10000	0.80
3500/12000	0.75

By research, inventor finds the relation of thermal resistance under different condition:

In the situation that having fan, thermal resistance Rth-ha is the function of air velocity;

In the situation that having fan, thermal resistance Rth-ha is the function of height above sea level;

In the situation that there is no fan, thermal resistance Rth-ha is the function of power loss;

In the situation that there is no fan, thermal resistance Rth-ha is the function of height above sea level.

Fig. 2 is illustrated in place, sea level, that is, air pressure is in 1 atmospheric situation, in the cooling system with fan, and thermal resistance and the relation of air velocity that flows through radiator.As can be seen from Figure 2, when air velocity is 3.0m/s, at fan, in normal operating conditions or radiator, do not have when blocked, thermal resistance value is approximately 0.4 ℃/W, and along with fan is aging or radiator by obstructions such as dusts, thermal resistance value increase, and when air velocity is 0.5m/s, thermal resistance value is approximately 0.9 ℃/W, and this just shows that cooling system goes wrong, and needs safeguard in time or change.

Fig. 3 is illustrated in different altitude height place, in the cooling system with fan, and the curve of the relation of thermal resistance and air velocity.As shown in Figure 3, when height above sea level is 2500m, air pressure is 0.737 atmospheric pressure, now, at fan under normal operating conditions, when air velocity is 3m/s, thermal resistance value is 0.46 ℃/W, the a little higher than thermal resistance value at place, sea level, that is to say, the thermal resistance value of now measuring need to be repaired by above-mentioned reduction factor.

Fig. 4 is illustrated in place, sea level, the relation of thermal resistance and power loss in the cooling system that there is no fan.As can be seen from Figure 4, in radiator lower time in shape, thermal resistance value is 1.4 ℃/W, and when radiator stops up, thermal resistance value is 3 ℃/W.

As can be seen from the above analysis; by according to the thermal resistance value of cooling system; can diagnose whole cooling system; that is to say judge whether the thermal resistance value of current cooling system exceedes a predetermined value, if exceeded; determine and need to safeguard or change cooling system; thus, user can preset equipment downtime according to diagnostic result, and cooling system is carried out to maintenance and repair.

But, because thermal resistance value is difficult to directly measure, therefore, in No. 201010127005.2nd, the common a Chinese invention patent application of submitting to of the applicant, a kind of method that reflects the thermal resistance value of current system by means of measurement temperature has been proposed.Below with reference to Fig. 5-7, describe the method in detail.

Fig. 5 is the hot loop figure of the power model cooling system of frequency converter, wherein, with reference to the analytical method of RC circuit, from the chip of power model, to the hot loop of radiator, can be equivalent to Fig. 5 B and 5C.Wherein Rth represents thermal resistance, and Zth represents thermal impedance.

As shown in Figure 5, in this model, heat from chip in series the interface of interface, base plate and the radiator of interface, circuit board and the base plate of the wiring copper layer of interface, scolder and the circuit board by chip and scolder (in the situation that of the sandwiched thermal grease conduction in centre, also comprise the interface of base plate and thermal grease conduction and the interface of thermal grease conduction and radiator) etc. flow to radiator, then from radiator, be dissipated to air.

For the response time of this model, for the power loss P of the step of an introducing, response time from chip to base plate is very fast, be approximately 100 microseconds, and response time from base plate to radiator is approximately a few minutes, but the response time of (air) is approximately several hours from radiator to surrounding environment, therefore the response time of surrounding environment can be thought to infinity, the temperature T a of surrounding environment can be thought to an invariable constant basis.

Regard power model as an entirety, whole model simplification is for as shown in Figure 5 C.

Be similar to Ohm's law, U=R × I, U is similar to the Δ Tca of native system, R is similar to the Rthca of native system, and I is similar to the P (power loss) in native system, if apply a current step to RC circuit, the voltage U at capacitor C two ends will raise gradually, and final voltage will meet above-mentioned Ohm's law.Model class of the present invention is similar to said system.

Above-mentioned model is carried out to modeling, to obtain the real-time status equation of this model:

P(t)＝P _Rth(t)+P _Cth(t) (1)

Be similar to the real-time analysis equation of RC circuit:

I(t)＝I _r(t)+I _c(t)

I _r(t)＝U(t)/R

I _c(t)＝C·d/dt(U(t))

Above-mentioned formula (1) can be derived into:

$P\left(t\right)=\frac{\mathrm{\&Delta;}{T}_{\mathrm{ca}}\left(t\right)}{\mathrm{Rt}{h}_{\mathrm{ca}}}+\mathrm{Ct}{h}_{\mathrm{ca}}\&CenterDot;\frac{d}{\mathrm{dt}}\&CenterDot;\mathrm{\&Delta;}{T}_{\mathrm{ca}}\left(t\right)---\left(2\right)$

For the step signal of an introducing P, carry out Laplace transform, obtain:

$\frac{P\left(s\right)}{s}=\frac{\mathrm{\&Delta;}{T}_{\mathrm{ca}}\left(s\right)}{\mathrm{Rt}{h}_{\mathrm{ca}}}+\mathrm{Ct}{h}_{\mathrm{ca}}\&CenterDot;s\&CenterDot;\mathrm{\&Delta;}{T}_{\mathrm{ca}}\left(s\right)---\left(3\right)$

$\mathrm{\&Delta;}{T}_{\mathrm{ca}}\left(s\right)=\frac{\mathrm{Rt}{h}_{\mathrm{ca}}\&CenterDot;\frac{1}{\mathrm{Rt}{h}_{\mathrm{ca}}\&CenterDot;\mathrm{Ct}{h}_{\mathrm{ca}}}}{s\&CenterDot;(s+\frac{1}{\mathrm{Rt}{h}_{\mathrm{ca}}\&CenterDot;\mathrm{Ct}{h}_{\mathrm{ca}}})}\&CenterDot;P\left(s\right)---\left(4\right)$

As mentioned above, τ=Rth _cacth _ca

Inverse Laplace transform, obtains

$\mathrm{\&Delta;}{T}_{\mathrm{ca}}\left(t\right)=P_{\mathrm{AV}}\&CenterDot;\mathrm{Rt}{h}_{\mathrm{ca}}(1-e^{\left(\frac{-t}{\mathrm{Rt}{h}_{\mathrm{ca}}\&CenterDot;\mathrm{Ct}{h}_{\mathrm{ca}}}\right)})=P_{\mathrm{AV}}\&CenterDot;\mathrm{Rt}{h}_{\mathrm{ca}}\&CenterDot;(1-e^{\left(\frac{-t}{\mathrm{\&tau;}}\right)})---\left(5\right)$

From formula (5), can find out, under cooling system normal operation, by design radiator, can obtain desirable Rth, and the Cth of radiator be also known, therefore, as long as calculate power loss P, can obtain Δ Tca.

In addition, can be by

be defined as the thermal impedance Zth of whole cooling system _ca(t), obtain following formula (5-1):

ΔT _ca(t)＝Zth _ca(t)·P _ca(t) (5-1)

Conventionally, the power loss of frequency converter is to calculate like this, first, suppose: output current (current of electric) is sinusoidal wave and is symmetrical, and mean P WM (power width modulated) is sine wave.

For the current of electric of U phase and the index of modulation of PWM, equal:

$m_U\left(t\right)=\frac{1}{2}(1+M\sin ({\mathrm{\&omega;}}_{0}t)$

Wherein, I _pEAKit is maximum motor electric current I _u, it is to calculate from the current measurement value of three phase electric machine, ω ₀the angular frequency (function of stator motor frequency) of motor, and it is the phase shift of current of electric.Average power consumption in inverter is:

$P_{\mathrm{inv}\_\mathrm{AV}}\&cong;I_{\mathrm{PEAK}}(\frac{1}{\mathrm{\&pi;}}({\mathrm{<figure-callout\; id="0"\; label="V\; CE"\; filenames="HSA00000040449900081.png"\; state="\{\{state\}\}">V</figure-callout>}}_{\mathrm{CE}}+V_{\mathrm{DF}0})+\frac{2(E_{\mathrm{ON}}+E_{\mathrm{OFF}}+E_{\mathrm{QN}})}{\mathrm{\&pi;}{V}_N{I}_N}{V}_{\mathrm{BUS}}{f}_{\mathrm{SW}}+I_{\mathrm{PEAK}}\left(\frac{(r_{\mathrm{CE}}+r_D}{4}\right))$

Wherein:

V _cE0, V _dF0, r _cEand r _dit is the static parameter of IGBT and fly-wheel diode;

V _bUSit is the DC busbar voltage recording;

F _sWit is the switching frequency of inverter;

V _nand I _nrespectively rated voltage and the rated current of motor;

E _oN, E _oFFand E _qNfor I _nand V _n, the switching loss of IGBT.

Total average power consumption of frequency converter is:

P _AV＝P _{inv_AV}+P _{rec_AV}

Wherein, P _{rec_AV}the power loss of diode rectifier, in the case of the different short-circuit capacity of supply network, P _{rec_AV}approximately P _{inv_AV}10% to 20%.

Thereby, the P calculating through said method is updated in formula (5), can be in the hope of Δ Tca ideally.

Referring to Fig. 6, Fig. 6 illustrates the control chart of said method.As can be seen from Figure 6, first measure Ta and Tc, obtain actual Δ Tca, and simultaneously, according to formula (5), calculate desirable Δ Tca, desirable Δ Tca is compared with the Δ Tca of actual measurement, draw the poor of the two, and when the difference of the two exceedes a predetermined threshold, definite radiator now stops up or fan is aging or damage.

That is to say, due to now because radiator is blocked or because fan is aging, damage, the Rth of actual cooling system is increase compared with design Rth, therefore, the radiating efficiency of cooling system reduces, and actual Δ Tca is increased, and when increase exceedes predetermined threshold, prove that cooling system now need to safeguard or change.

Due in the frequency converter of prior art, be provided with temperature sensor and measured Tc, therefore, in order to realize said method, the power loss that a temperature sensor carrys out measures ambient temperature Ta and needs accurate Calculation frequency converter need to be set in addition.

For fear of increasing a temperature sensor, measure Ta, as shown in Figure 7, because the Ta mentioning in the above can regard a steady state value substantially as, therefore, to Δ, Tca differentiates, and can avoid like this increasing a temperature sensor and measure Ta.

Obtain following formula (5-2):

$\frac{d\&CenterDot;\mathrm{\&Delta;}{T}_{\mathrm{ca}}\left(t\right)}{\mathrm{dt}}=\frac{d\&CenterDot;T_c\left(t\right)}{\mathrm{dt}}=\frac{P_{\mathrm{AV}}}{\mathrm{Ct}{h}_{\mathrm{ca}}}\&CenterDot;e^{\left(\frac{-t}{\mathrm{\&tau;}}\right)}---(5-2)$

By such improvement, can avoid increasing a temperature sensor and carry out measures ambient temperature Ta, and increase thus the robustness of system.

But, in the applicant's said method, can find out formula (5) or (5-2) in all comprise power loss P _aV, and this just need to accurately calculate or measure power loss, and as mentioned above, accurate Calculation power loss P _aVconventionally need very intensive, and be sometimes difficult to guarantee precision.

For this reason, according to the present invention, provide execution mode as described below, solved the problems referred to above and improve the robustness of system.

preferred implementation

The formula (5) of obtaining during as above in the face of system modelling:

$\mathrm{\&Delta;}{T}_{\mathrm{ca}}\left(t\right)=P_{\mathrm{AV}}\&CenterDot;\mathrm{Rt}{h}_{\mathrm{ca}}(1-e^{\left(\frac{-t}{\mathrm{Rt}{h}_{\mathrm{ca}}\&CenterDot;\mathrm{Ct}{h}_{\mathrm{ca}}}\right)})=P_{\mathrm{AV}}\&CenterDot;\mathrm{Rt}{h}_{\mathrm{ca}}\&CenterDot;(1-e^{\left(\frac{-t}{\mathrm{\&tau;}}\right)})---\left(5\right)$

Above-mentioned formula (5) is differentiated, and as mentioned above, Ta is thought invariable, obtain formula (6)

$\frac{d\&CenterDot;\mathrm{\&Delta;}{T}_{\mathrm{ca}}\left(t\right)}{\mathrm{dt}}=\frac{d\&CenterDot;T_c\left(t\right)}{\mathrm{dt}}=\frac{P_{\mathrm{AV}}}{\mathrm{Ct}{h}_{\mathrm{ca}}}\&CenterDot;e^{\left(\frac{-t}{\mathrm{Rt}{h}_{\mathrm{ca}}\&CenterDot;\mathrm{Ct}{h}_{\mathrm{ca}}}\right)}---\left(6\right)$

Wherein:

p _aV, Cth _caand Rth _cafunction,

And Cth _caconstant and known; Rth _caby the function of the air velocity of radiator.

In order to eliminate power loss P, in system, in stable state in the situation that, introduce a step signal of power loss P, the step signal of power loss can or change load by for example switching frequency and realize.In order to eliminate the impact of power loss P, introduce a gain k simultaneously ₀, and by this gain k ₀be defined as:

${\mathrm{<figure-callout\; id="0"\; label="k"\; filenames="HSA00000040449900081.png"\; state="\{\{state\}\}">k</figure-callout>}}_0=\frac{d\&CenterDot;T_c\left(0\right)}{\mathrm{dt}}=\frac{P_{\mathrm{AV}}}{\mathrm{Ct}{h}_{\mathrm{ca}}}$

T=0 is very short a bit of time after introducing power loss step just;

Because system is in stable state, therefore P _aVafter t=0 and during all detections, be constant, thus, at all detection time k ₀remain unchanged, this can limit by the motor control with constant power loss.

Use k ₀except dTc/dt, obtain formula (7),

$\frac{d\&CenterDot;T_c\left(t\right)}{\mathrm{dt}}\&CenterDot;\frac{1}{{\mathrm{<figure-callout\; id="0"\; label="k"\; filenames="HSA00000040449900081.png"\; state="\{\{state\}\}">k</figure-callout>}}_0}=e^{\left(\frac{-t}{\mathrm{Rt}{h}_{\mathrm{ca}}\&CenterDot;\mathrm{Ct}{h}_{\mathrm{ca}}}\right)}---\left(7\right)$

This shows, formula (7) is and P _aVirrelevant function, according to formula (7), obtains control chart as shown in Figure 8, and this control chart illustrates diagnostic method according to the preferred embodiment of the present invention.

Preferably, in order to obtain better robustness, detection time is defined as $t=\frac{{\mathrm{\&tau;}}_{\mathrm{ca}}}{4}$ Or larger, formula (7) can be derived into:

$\frac{d\&CenterDot;T_c\left(\frac{{\mathrm{\&tau;}}_{\mathrm{ca}}}{4}\right)}{\mathrm{dt}}\&CenterDot;\frac{1}{{\mathrm{<figure-callout\; id="0"\; label="k"\; filenames="HSA00000040449900081.png"\; state="\{\{state\}\}">k</figure-callout>}}_0}=e^{\left(\frac{-K_d}{4}\right)}---(7-1)$

Wherein: K _dbe a fiducial value, and can provide a control table for different t, to compare with measured value.And for robustness better, K _dpreferably equal 0.5, that is to say, when the timeconstantτ of measuring _cathe τ of design _catwice time, result of detection is 1, means that fan failure or radiator stop up.

Fig. 9 and Figure 10 show the robustness of diagnostic method according to the preferred embodiment of the present invention.Wherein abscissa is time t.

As shown in Figure 9, three curves shown in Fig. 9, a bottom actual measured value that curve a is dT/dt, a middle curve b is under normal circumstances, i.e. dT/dt value under cooling system normal operation, and one of the top curve c is by introducing k ₀and the curve of the impact of removal power loss, by these three curves, can be found out, difference between curve a and curve c diminished gradually along with the time, therefore cannot whether have problems by Accurate Diagnosis cooling system, and can find out from curve b and curve c, in the in-problem situation of cooling system, the gap of the two is larger, therefore, can determine reliably cooling system existing problems.

Equally, as shown in figure 10, a bottom curve a is illustrated in the dT/dt value under cooling system normal operation, and middle curve b is illustrated in the dT/dt value under fan aging conditions, and topmost a curve c is illustrated in the dT/dt under fan damaged condition.From three curves of Figure 10, can find out, along with passage of time, difference between curve b and c and curve a is increasing, therefore, according to current measurement value (after processing), (curve a) relatively, if actual measured value is greater than threshold value, illustrates that cooling system has problems with threshold value, and can judge the residing situation of cooling system according to the difference of the two, as aging or damage completely.

As can be seen from Figure 10, detection time more after, survey robustness better, preferably, $t=\frac{{\mathrm{\&tau;}}_{\mathrm{ca}}}{4}$ Or larger.

Above, by preferred implementation above, the present invention has been described, this shows, in diagnostic method of the present invention, avoided accurate Calculation or measured power loss, reliability and the robustness of diagnosis have been improved thus, thus, the present invention has realized the reliable diagnosis to cooling system, and makes user in the situation that not increasing any cost, to cooling system, to carry out real-time diagnosis, to find in time the problem of cooling system, and carry out planned maintenance and repair.

Claims (7)

1. for a diagnostic method for the cooling system of electrical system, this diagnostic method is characterised in that and comprises the following steps:

At this electric equipment, under stable state, to it, introduce a step signal;

Introducing after this step signal, measuring the temperature of the heater element of described electric equipment;

Obtained temperature value is carried out to signal processing; And

Result after treatment signal is compared with predetermined threshold, and when described result is greater than described threshold value, judge that described cooling system work is undesired, wherein

Described stable state refers to that described electric equipment is operated under stable load and switching frequency, maintains stable working temperature; And

The step of obtained temperature value being carried out to signal processing comprises:

To obtained temperature value seeking time derivative; And

By obtained time-derivative by a coefficient k ₀remove this coefficient k ₀the derivative value of above-mentioned measured temperature during for time t=0, that is:

Wherein: P _aVfor power loss; And Cth _cafor the thermal capacitance of cooling system.

2. the method for claim 1, wherein described cooling system work is undesired comprises that the fan of described cooling system is aging, and fan damage or radiator are by foreign matters from being blocked.

3. method as claimed in claim 1 or 2, wherein, described step signal is power loss step signal.

4. method as claimed in claim 3, wherein, described power loss step signal is the switching frequency by changing described electric equipment or changes load or change and realize simultaneously.

5. the method for claim 1, wherein described threshold value is:

Wherein: Rth _cafor the thermal resistance value of system under normal operation; And t is the testing time.

6. method as claimed in claim 5, wherein, described testing time t is

or larger, wherein, the thermal time constant that τ is system.

7. the method for claim 1, wherein described electric equipment is frequency converter, and the power model that described heater element is frequency converter.

Images